RAG on Me .
A résumé you could ask questions.
Concept illustration / A view of the idea, not a product screenshot.
01 / The context
The problem
worth working on.
A résumé is a compressed view of someone’s work. I wanted recruiters to ask natural questions and receive answers grounded in the actual documents.
02 / The approach
How the pieces
come together.
- 01
Ingested résumé and project documents into a vector store and connected retrieval to a conversational graph.
- 02
Used persistent conversation state to keep follow-up questions in context.
- 03
Exposed the assistant through a FastAPI service and the original portfolio chat interface.
03 / A key decision
The detail
that shapes the system.
Constrain the assistant to documented experience. A persuasive answer is not useful if it invents skills or achievements.
04 / The result
What came out of it.
An early hands-on RAG project connecting ingestion, retrieval, generation, and persistent memory.
Scope & limitations
The original hosted chatbot is retired. The source remains available as project history; a replacement assistant is not live on this portfolio.